tooluniverse-functional-genomics-screens

Validate each hit against DepMap dependency scores and gnomAD constraint metrics, then run Reactome pathway enrichment and STRING network analysis. Score hits on essentiality selectivity, pathway convergence, druggability, clinical evidence, and literature support to produce tiered rankings.

Use the bundled depmap_gene_dependency.py script, which downloads the DepMap Public CRISPRGeneEffect.csv once and queries by gene or cell line. Chronos scores below -0.5 indicate dependency and below -1.0 indicate strong essentiality.

DGIdb provides drug-gene interactions and druggability categories such as kinases and GPCRs. For high-priority hits, clinical trial searches and PubMed queries find novel inhibitors not yet cataloged in DGIdb.

It works but with caveats, since DepMap covers roughly 1000 cancer cell lines and is less informative for non-cancer contexts. For non-cancer screens, weight gnomAD constraint and pathway evidence more heavily than DepMap essentiality.

The workflow does not perform raw screen analysis, which requires upstream tools like MAGeCK or BAGEL. STRING interactions represent associations rather than causal relationships, and shRNA screens need a higher validation bar due to off-target effects.